ASTROPLANE best vertical antenna ever?

[nosepc]

hello

Hello again.

I find all this very interesting discussion thread.

And what I'm taking in conclusion ...

is that the other "super" famous vertical antennasnot as good or better, but rather commercial properties are myths.??

Not in actual field tests, or models are large differences in performance, but rather are negligible.

We found that almost all commercial pamphlets are just that.?

greetings from Argentina

nosepc

 

[The DB]

There are several factors that determines how a vertical antenna performs.

The best thing you can do with any given vertical antenna is make it higher. The higher it is the better it will perform, unless you have some way of getting extreme heights you will not be able to put it to high.

Second comes the quality of the ground plane. Beware any numbers that assume a "perfect" ground plane, such a thing does not exist, and the only thing that comes close on the planet is a salt water environment. Also, don't assume that four elevated radials comes close to simulating a "perfect" ground plane, they don't...

Third comes the length of the antenna. In the real world a 1/2 wavelength and a 5/8 wavelength antenna are not as far performance wise as the "3dB gain" figure often cited as the difference between the two. This is an example of a number using a "perfect" ground plane as a reference. Perhaps if you mount the antennas just above a salt water environment you will see that, but not elsewhere.


The DB

 

[Marconi]

There are several factors that determines how a vertical antenna performs.

The best thing you can do with any given vertical antenna is make it higher. The higher it is the better it will perform, unless you have some way of getting extreme heights you will not be able to put it to high.

Second comes the quality of the ground plane. Beware any numbers that assume a "perfect" ground plane, such a thing does not exist, and the only thing that comes close on the planet is a salt water environment. Also, don't assume that four elevated radials comes close to simulating a "perfect" ground plane, they don't...

Third comes the length of the antenna. In the real world a 1/2 wavelength and a 5/8 wavelength antenna are not as far performance wise as the "3dB gain" figure often cited as the difference between the two. This is an example of a number using a "perfect" ground plane as a reference. Perhaps if you mount the antennas just above a salt water environment you will see that, but not elsewhere.


The DB

DB, I tend to agree with your post for the most part, but do you remember this thread and our discussions on the 5/8 wave Mystique? http://www.worldwidedx.com/cb-antennas/154101-5-8-wavelength-antenna-mystique.html

If we consider this work as a defacto reference to antenna gain, then I don't see the 5/8 wave antenna showing any where near 3db gain over a 1/2 wave...even over an infinite ground plane.

Here is another published example for antenna gain..."The Truth Table about CB antennas" by Orr and Cowan. It too shows about the same...a 1.2db gain for the 5/8 wave over the 1/2 wave.

I use to believe the idea that a 5/8 wave antenna was able to produce a signal that was double the gain of a 1/4 wave ground plane too, until I tested them and found out the difference in performance was not as claimed in almost ever CB discussion on the subject, or else the differences noted were misunderstood and/or taken out of context, while the facts associated with the testing were also ignored. I said to myself, this is a real recipe for CB BS.

I consider the latter to be true here, and manufactures have for ever taken liberties with the truth as noted in Ballentine's study in the early 1920's. For me this is also why Reynold's wrote his critique of Ballentine's work...he saw the wickedness created by all the lies flourishing in CB world of the 1970-1980's, and it continues up to today.

I also believe that both Cebik and W8JI have debunked, to some degree, the ideas that even the differences noted here, between a 1/4 wave and a 5/8 wave, are not as much as indicated...when placed in real world situations and when other losses are considered.

 

[The DB]

If we consider this work as a defacto reference to antenna gain, then I don't see the 5/8 wave antenna showing any where near 3db gain over a 1/2 wave...even over an infinite ground plane.

I did some playing in 4nec2 with a 1/4 1/2 and 5/8 wavelength radiators over a "perfect" ground...

We have the 1/4 wavelength, the 1/2 wavelength, and the 5/8 wavelength models in that order left to right.

These three models are a vertical radiator with the bottom of the element connected to a "perfect" ground as the software lets me connect them if I wish. There are no radials for a groundplane as the groundplane already exists in the "perfect" ground. There are 100 segments per 1/4 wavelength with the feedpoint being the first segment above the "perfect" ground.

I'm seeing a difference in gain of 2.97 dB between a 1/4 and 5/8 wavelength antennas, which is close enough for most people to round to 3 dB.

Conclusion, it is possible to achieve a figure near the mythical 3 dBq gain figures when dealing with only a radiator over "perfect" ground. But as a "perfect" ground does not exist in the real world the gain figure is irrelevant.

Now if your talking a antennas with four 1/4 wavelength radials, even if it is near said "perfect" ground plane I can't seem to get anywhere near a 3 dBq gain when using a 5/8 wavelength vertical.

I played further with various other configurations and ground qualities and was unable to match the 3 dBq gain as shown above.

Of course, all of this is assuming I did not make a mistake somewhere...


The DB

 

[Marconi]

I agree, your models are close enough for me, and they suggest the same as the references I used. Was that a surprise?

I read your comment that you meant to compare a 1/2 wave to a 5/8 wave. That is why I said I agree with you in part. We all mis-state our thoughts at times.

 

[The DB]

I mis-typed when I said 1/2 wavelength. Not all bad though, I got to play with modeling some more, and that is something I want to do more of in general... I still have a ton to learn when it comes to modeling...


The DB

 

[W5LZ]

Antenna modeling programs are very interesting! They can shed light on all sorts of things, show you the relationships between various antennas in particular situations. They are also only an 'indication', never a definite 'answer' to things because they typically only deal with the information you 'feed' them. If you can input the characteristics of the antenna and everything around them they would be closer to 'real world' results. Good luck with that! And because entering all that information is very unlikely (impossible?) the results are still only 'theoretical', not 'real'.
They are still fun, aren't they?
- 'Doc

 

[Marconi]

I mis-typed when I said 1/2 wavelength. Not all bad though, I got to play with modeling some more, and that is something I want to do more of in general... I still have a ton to learn when it comes to modeling...

The DB

DB, I hope you to continue with your modeling. I don't have it all in hand either...by a long shot, and it would be nice to have someone to talk with in that regard.

Could you answer these questions for me? This is not a test, I would like to know because I have not mastered 4nec2 well enough to feel good yet.

Do you see anyway for 4nec2 to display the actual currents and phase per segment...similar to how my Eznec's tabular currents log does?

In your description process using an editor, can you attach two wires together as follows: wire #1 end 1 to wire #2 end 1 without getting an error alert?

Does your program provide a feature that creates radials automatically, or do you have to construct radials one wire at a time?

Eznec always shows currents turned off, but has a switch that allows currents to be turned on. Does 4nec2 have a similar feature?

 

[nosepc]

hello.

I find it much easier didactic, complete the 4nec2.

EZNEC do not like, I find it bizarre and little graphic...

 

[The DB]

This is not a test

Always a good start... No pressure on me then...

Do you see anyway for 4nec2 to display the actual currents and phase per segment...similar to how my Eznec's tabular currents log does?

In the .out file there is a section that has data that seems to have what you are asking about...

  SEG.  TAG    COORD. OF SEG. CENTER     SEG.            - - - CURRENT (AMPS) - - -
  NO.   NO.     X        Y        Z      LENGTH     REAL        IMAG.       MAG.        PHASE
     1    1   0.0000   0.0000   0.0122  0.00623   3.1315E-04  1.8179E-03  1.8446E-03   80.226
     2    1   0.0000   0.0000   0.0184  0.00623   3.1207E-04  1.7391E-03  1.7668E-03   79.827
     3    1   0.0000   0.0000   0.0247  0.00623   3.1070E-04  1.6529E-03  1.6818E-03   79.354
     4    1   0.0000   0.0000   0.0309  0.00623   3.0907E-04  1.5684E-03  1.5986E-03   78.852
     5    1   0.0000   0.0000   0.0371  0.00623   3.0719E-04  1.4839E-03  1.5153E-03   78.304
     6    1   0.0000   0.0000   0.0434  0.00623   3.0505E-04  1.3987E-03  1.4316E-03   77.697
     7    1   0.0000   0.0000   0.0496  0.00623   3.0265E-04  1.3128E-03  1.3472E-03   77.018
     8    1   0.0000   0.0000   0.0558  0.00623   3.0000E-04  1.2258E-03  1.2620E-03   76.248
     9    1   0.0000   0.0000   0.0621  0.00623   2.9708E-04  1.1379E-03  1.1760E-03   75.367
    10    1   0.0000   0.0000   0.0683  0.00623   2.9390E-04  1.0489E-03  1.0893E-03   74.347

That is just segments 1 to 10 of the 100 segments in that wire, and there are four additional wires with 100 segments each so I settled for a sample... Also there is a scroll bar to scroll left and right, the phase is on the far right...

In your description process using an editor, can you attach two wires together as follows: wire #1 end 1 to wire #2 end 1 without getting an error alert?

I have done so multiple times so far and have yet to get such an error message.

Does your program provide a feature that creates radials automatically, or do you have to construct radials one wire at a time?

I have done them one wire at a time, I have not yet seen a feature to create radials automatically, but I haven't looked yet either.

Eznec always shows currents turned off, but has a switch that allows currents to be turned on. Does 4nec2 have a similar feature?

Can you show me an example of what your referring to here? I think I know but want to be sure.


The DB

 

[Marconi]

In the antenna view, here is how Eznec shows currents switched off and on.





Maybe 4nec2 always has currents turned on from the start. I think Exnec starts in the off postition. I don't know why one even needs currents to be turned off.

 

[The DB]

Does your program provide a feature that creates radials automatically, or do you have to construct radials one wire at a time?

I have found a feature of the program to create radials automatically, however every time they have been created thus far they have been on/in ground. I have yet to figure out how to create elevated radials automatically with 4nec2.

Eznec always shows currents turned off, but has a switch that allows currents to be turned on. Does 4nec2 have a similar feature?

Was just making sure on this, and yes 4nec2 does this as well.

And they can also be displayed like this...

Further, and not shown, I can display current and phase together in several ways, for example the top picture in this post can show phase by the current data line crossing the modeled antenna. I can also display current and phase separately on the same model, for example, current as the first picture and phase as the second picture, or vice versa.

So the short answer here is yes.

The above pictures is from a 5/8 wavelength vertical test model I used when playing around today.

---

Speaking of playing around with 4nec2, I did some test models at various heights using 1/4 and 5/8 wavelength verticals, both with four 1/4 wavelength horizontal radials. This is testing some things I and others said earlier in this thread.

I ran the 1/4 and 5/8 wavelength models a little differently from each other, I would tune the vertical length of the quarter wavelength antenna with the top hats of various sizes to have a resonance match near the middle of the CB band. All of the tests were over 4nec2's real "Moderate" ground at 1/4, 1/2, and 1 wavelengths above ground to the base of the antenna. For the purposes of these tests I did not include a mast.

Using 1/4 wavelength vertical and top hats I adjusted the vertical length of said antennas to keep resonance near the middle of the CB band. I did this for consistency of tuning. Predictably the larger the cap hat the smaller the vertical element of the antenna needed to be. I was surprised at how quickly the vertical length dropped with the increase in cap hat size. I was also surprised at the radiation patterns between the various antennas mounted at given heights were all the same after their heights were adjusted for resonance. I previously made this statement in this thread...

That is fine, except physically shortening the antenna has the effect of making the horizontal lobes wider (by wider I am referring to more signal going further up and down rather than straight out), thus not as much energy is being directed out, this lowers maximum gain.

This statement seems to be in error, so I stand corrected. The physical shortening effect of a cap hat does not seem to change the radiation pattern of the shorter antenna as I expected it to. However, the shortened 1/4 wavelength antenna using a cap hat still has less overall gain than a longer full length version.

Using the 5/8 wavelength models I mentioned above, I ran these a little differently. As I have yet to play with any antenna matching short of actually changing the antenna itself I was unable to match the various 5/8 wavelength models I tested, so I kept them all at 5/8 wavelength in overall vertical length. These results, however, are right in line and expected based on the results of the 1/4 wavelength tests above.

Just as above the cap hats seem to simulate additional overall physical length when comparing radiation patterns. Being that I started with 5/8 wavelength vertical antennas, as the cap hat sizes increased the high angle lobe I expected to see developing developed and very quickly. The cap hats didn't appear to, as Wavrider put it:

The top hat seemed for lack of a better term to "force" the radiation down to the horizon per say.

I was unable to duplicate the effect he mentioned. However, to be fair, I am not satisfied with my modeling skills to say definitively that he was incorrect. I questioned this statement before and cited other possibilities I noticed in the models that could potentially cause the same effect to occur. I am still not convinced one way or the other, but with what I've seen I currently disagree with the premise of Wavrider's statement.

Now if only I saved the various models I made to post up some results instead of just changing the data in the same file... Perhaps at some point I'll go through creating said models again if anyone is interested in seeing the actual result data, but not tonight...


The DB

 

[Marconi]

I have found a feature of the program to create radials automatically, however every time they have been created thus far they have been on/in ground. I have yet to figure out how to create elevated radials automatically with 4nec2.



Was just making sure on this, and yes 4nec2 does this as well.

And they can also be displayed like this...

Further, and not shown, I can display current and phase together in several ways, for example the top picture in this post can show phase by the current data line crossing the modeled antenna. I can also display current and phase separately on the same model, for example, current as the first picture and phase as the second picture, or vice versa.

So the short answer here is yes.

The above pictures is from a 5/8 wavelength vertical test model I used when playing around today.

Well it is good to know that 4nec2 also does what Eznec does in this regard and starts models with the currents turned off.

Just as a note, in the past while talking to Shockwave about his ideas on the S4/Vector and Eznec...I thought this topic of starting a model with currents turned off...might be part of the problem he was describing, IE., referencing your model above that does not seem to indicate the bottom of the 5/8 wave radiator being out of phase with the top 1/2 wave portion as we would expect. IMO the ideas noted here are very important to understand modeling if one gets into details regarding currents and phase.

For me it took about 2 years before I even considered the differences you've noted for how your 4nec2 can report out different looks concerning currents and phase. To me this understanding is vitally important if we are to attempt to try and prove or explain how an antenna really works...at least with respect to the level of detail that Shockwave and Bob, among others, like to discuss.

For me however, this subject of currents and phase is just part of a leaning curve. I don't have things as easy as some do in this area, so I don't try and get all categorical on the issues...even though I might try and talk about the subject as I understand it. I've been trying for a long time now to try and find someone that would talk about currents and phase or try and learn along with me. I've even tried to talk to Roy Lewallen about how currents work in Eznec, but that was to noavail. I tried to talk to Henry in Holland, but that went down a rabbit hole too. I did not understand a thing he was trying desperately to explain to me. But that could have just as easily been all my fault.

---

Speaking of playing around with 4nec2, I did some test models at various heights using 1/4 and 5/8 wavelength verticals, both with four 1/4 wavelength horizontal radials. This is testing some things I and others said earlier in this thread.

I ran the 1/4 and 5/8 wavelength models a little differently from each other, I would tune the vertical length of the quarter wavelength antenna with the top hats of various sizes to have a resonance match near the middle of the CB band. All of the tests were over 4nec2's real "Moderate" ground at 1/4, 1/2, and 1 wavelengths above ground to the base of the antenna. For the purposes of these tests I did not include a mast.

Using 1/4 wavelength vertical and top hats I adjusted the vertical length of said antennas to keep resonance near the middle of the CB band. I did this for consistency of tuning. Predictably the larger the cap hat the smaller the vertical element of the antenna needed to be. I was surprised at how quickly the vertical length dropped with the increase in cap hat size. I was also surprised at the radiation patterns between the various antennas mounted at given heights were all the same after their heights were adjusted for resonance. I previously made this statement in this thread...

This statement seems to be in error, so I stand corrected. The physical shortening effect of a cap hat does not seem to change the radiation pattern of the shorter antenna as I expected it to. However, the shortened 1/4 wavelength antenna using a cap hat still has less overall gain than a longer full length version.

Using the 5/8 wavelength models I mentioned above, I ran these a little differently. As I have yet to play with any antenna matching short of actually changing the antenna itself I was unable to match the various 5/8 wavelength models I tested, so I kept them all at 5/8 wavelength in overall vertical length. These results, however, are right in line and expected based on the results of the 1/4 wavelength tests above.

DB, I have the same problems with all antenna models I've made that require a matcher. However, for the most part our ignoring the match, as it were, is not as important as it might appear. You can look until you're blue in the face and hardly ever find an example of antennas requiring matching with a matching device included in the model. The few I have found, do not show me a good match when I run the source output of the antenna anyway, so I'm still in the dark about what that is all about.

I can and have built to specs the physical matching device (trombone) for my I-10K, and I can get the match to work for the model to effectively change the match, but I'm not sure of such results, because with these serial type inductors added...the model produces a pattern that seems to always show the antenna producing both horizontal and vertical polarity patterns.

At any rate I don't see there being much difference in the gain and angles noted for the antenna with a 1.20 SWR or a 15.5 SWR, so maybe that is the point as to our not needing to be overly concerned. Plus I generally see my models showing "less than" rather than "more than" good results as the model improves, but I could be wrong on that too.

Just as above the cap hats seem to simulate additional overall physical length when comparing radiation patterns. Being that I started with 5/8 wavelength vertical antennas, as the cap hat sizes increased the high angle lobe I expected to see developing developed and very quickly. The cap hats didn't appear to, as Wavrider put it:

I was unable to duplicate the effect he mentioned. However, to be fair, I am not satisfied with my modeling skills to say definitively that he was incorrect. I questioned this statement before and cited other possibilities I noticed in the models that could potentially cause the same effect to occur. I am still not convinced one way or the other, but with what I've seen I currently disagree with the premise of Wavrider's statement.

Now if only I saved the various models I made to post up some results instead of just changing the data in the same file... Perhaps at some point I'll go through creating said models again if anyone is interested in seeing the actual result data, but not tonight...

The DB

I'll wait for more information if you try and duplicate the rest of your post regarding the ideas that you and Wavrider were discussing, but I think I might agree with him if the models are not really getting up high enough to really start showing other lobes being produced as we might expect, maybe up to about 1.2 wavelengths above the ground.

When we deal with top hat effectiveness we are likely discussing height as a matter of fractions of a wavelength like we find in very low frequency ham operations with very tall ground mounted antennas, and not actual height in feet about Earth like we tend to do in CB.

Thanks DB for the help and the discussion about 4nec2.

Thank you too nospec. I've learned something new today, and that is always good.

 

[Marconi]

I have found a feature of the program to create radials automatically, however every time they have been created thus far they have been on/in ground. I have yet to figure out how to create elevated radials automatically with 4nec2.

Try making one radial where you want it on the antenna, and then use the automatic radial making feature using the radial as a templet.

 

[The DB]

Nope, still puts it on the ground. The tool is called ground screen and I think it is meant to simulate radials on or under the ground on a ground mounted antenna.


The DB